The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard development commands like git, gh, npm, and pytest. These are executed within the context of issue resolution and are typical for development tools.
[DYNAMIC_CONTEXT_INJECTION]: The skill uses !command syntax in the Context section of SKILL.md to display git status, git branch, gh issue list, etc. These are safe, information-gathering commands used to provide the agent with local repository state.
[REMOTE_CODE_EXECUTION]: None detected. The quality validation commands (npm test, pytest, etc.) are standard local testing tools.
[DATA_EXFILTRATION]: No exfiltration patterns found. The skill interacts with GitHub through the official gh CLI for issue and PR management.
[PROMPT_INJECTION]: No malicious injection patterns found. The instructions use natural language to define a TDD and PR workflow.
[CREDENTIALS_UNSAFE]: No hardcoded credentials detected. The skill uses gh auth status to check existing authentication state.
[OBFUSCATION]: No obfuscated content, Base64, or hidden URLs found.

resolve-issues